The release of new generative platforms underscores the disruptive potential of artificial intelligence.
- Until recently, machines have been confined to analytical work, but artificial intelligence (AI) use cases are expanding owing to improved models, access to new data sets, and increasingly powerful hardware.
- Generative AI systems are better equipped to produce original content—including images, text, and speech—which is expected to have wide-ranging application.
- While generative AI has received the most attention recently, AI is being embedded into enterprise applications to improve products and services across a wide range of industries.
- In addition to creating value, AI has the potential to enhance and reshape the global economy in unpredictable ways and is likely to be highly impactful to the investments we make on behalf of our clients.
Artificial intelligence (AI) has the potential to transform society with the same force as the internet and smartphones have over the last two decades. Drawing analogies from these two innovations, we anticipate AI will be deeply transformative for almost every business, in ways that are difficult to imagine today. Even though many implications of AI are unknown today, professional services firm PwC estimates that AI could contribute up to $15.7 trillion to the global economy in 2030—with $6.6 trillion from increased productivity and $9.1 trillion from consumption-side effects.1
AI has been making steady progress over the last decade owing to advances in computing and better access to quality data sets. More recently, however, OpenAI’s ChatGPT chatbot model has captured the public’s attention with its ability to answer, augment and automate across a wide range of knowledge-based functions. While far from fool proof, the model has been known to pass qualifying exams in several fields and could be a clarion call for how education, and society more broadly, must prepare for a new wave of disruptive innovation.
Knowledge workers are not the only ones who have taken notice. DALL-E, the artistic form of OpenAI, has also made waves with its creative expression in poetry and imagery. By applying the principle of scaling computing resources to an ever-expanding set of data, society is poised to see a dramatic shift and acceleration in innovation as artistic and scientific fields embrace this platform advancement.
We expect that sustained interest and investment in AI will drive a new wave of growth in the coming decade as large-language models grow in both size and sophistication. As consumer dollars become more difficult to capture and data regulations make existing means of data capture and usage more difficult, AI may become the most relevant means of gaining a competitive advantage. With every passing year, the hardware supporting AI models is likely to become more powerful. Starting with texts, images, and simple videos, AI output would eventually progress to more closely mimic human intelligence, and potentially beyond. The belief that AI is driving the marginal cost of intelligence towards zero is becoming a reality.
How does AI work?
AI models like GPT-3 are modelled to function like the human brain. They feature software algorithms that calculate answers in deep neural networks on purpose-built hardware. The algorithms are constructed by training them to predict answers through iterative tests from a large sample set of training data that fine-tune the model’s ability to recognise patterns.
What can AI achieve?
AI systems can now intelligently process and mimic human speech by querying databases and delivering human-like responses while considering linguistic nuances such as tone and sentiment. This could have applications in automated helpdesks and call centres, search engines, and teaching. At a more personal level, natural language processing can deliver more human-like AI assistance beyond what digital voice assistants can do today. It has the potential to help humans access knowledge beyond what the internet can achieve via text search, or study individual writing styles and craft emails using just a few key words. AI companions could be spawned, as would personalities that only exist in the metaverse.
AI also has an unlimited potential for the creative industry by composing songs, making art, and, in the future, videos. Text-to-image use cases may still be rudimentary today, but in a few years could transform the creative industry. AI also holds significant potential in music given the mathematical construct behind rhythms, melodies, and styles. When coupled with natural-language processing, future hits may be created by unnamed AI systems rather than humans. Indeed, there have already been several chart-topping songs created by AI.
While advances in generative AI have made recent headlines, progress in other applications also continues to be robust.
Patterns and discoveries
Traditionally intensive analytical tasks requiring pattern analysis and predictive capabilities can become much more achievable and repeatable at scale with AI. Protein folding is one such domain, given its mathematical complexities. AI systems can sift through troves of data to find correlations and draw inferences, an undertaking that a human or computer today may find too complex. Some outputs include solving complex climate-change problems, spotting crime, generating computer code, and uncovering new biological findings. Gartner expects 50% of computer code to be generated by AI by 2026.
AI is able to process environmental inputs to make real-life decisions for tasks such as driving, cleaning, construction, delivery, repair, and making AI droids more cost-effective and safer. The intelligent home vacuum is the first simple application—drawing upon limited spatial information and learning about its environment through repeated encounters. Eventually, this technology could extend into more complex tasks in varied environments.
Although AI has received criticism owing to its potential to reflect societal biases, it is well-suited to offer positive contributions to environmental and social issues. For instance, by managing an enormous amount of data, it can assist in measuring Scope 3 emissions and potentially determine ways to lower them. According to another report from PwC, using AI for environmental applications has the potential to increase global GDP by 3.1% to 4.4% while lowering greenhouse-gas emissions by 1.5% to 4.0% by 2030. This is significant given how challenging it is to decouple economic growth from greenhouse-gas emissions.
From a human resources perspective, AI can uncover potential employees who may be unusually qualified. In addition to helping to find talented candidates, certain AI modes are well trained models to identify groups of present employees who run the danger of becoming disengaged.
Furthermore, AI and machine-learning systems are able to identify potential supply-chain hitches far in advance of material problems, an issue brought to light in the wake of the global pandemic. Supply-chain managers can use these tools to monitor risks like dangerous working conditions, human rights abuses, or the overall health of the supply-chain system. Incorporating environmental, social and governance (ESG) values in AI development may not only benefit businesses, but it could also benefit society if its risks are managed and these positive use cases come into fruition.
AI through a private-markets lens
Venture-capital (VC) investment in generative AI has seen a swift uptick since 2020 owing to advancements in technology and a growing recognition of its potential application in multiple industries. According to data from Pitchbook, there was just under $500 million invested into the industry in 2020. In 2022, over $2 billion flowed into the area, with multiple start-ups raising large sums of capital at high valuations from top VC firms and large tech companies.
VC firms are keeping a close eye on generative AI and have shared their outlooks on the area for 2023. Some are forecasting that there will be a shift from traditional analytical AI toward a focus on generative AI, with others predicting that early success for this technology will be seen in non-clinical operations. There does, however, seem to be ample opportunity to incorporate this emerging technology within the health-care ecosystem, especially when considering the recent launch of a $350 million fund focusing on the intersection of life sciences and computing. While some firms identified specific opportunities in back-office operations such as supply-chain management, procurement, and business-process outsourcing, others predict that generative AI will transform every industry that relies on original, human-generated output.
How will AI evolve from here?
We believe the first use cases will be consumer-oriented and recreational rather than within regulated domains and areas that are highly consequential on human life. Once these applications are established in areas such as text responses or simple droids for repetitive tasks, the next phase of more sophisticated use cases should follow.
Regulations are likely to be required to establish guard rails on how far AI can go. Will it be a winner-takes-all dynamic? We envisage AI systems would be implemented at various scales, from the largest models run at, or by, the ‘hyperscalers’ (those who can expand capacity to meet growing demand), to industry-specific models and individual ones. Smaller AI models may not be able to compete with the largest ones, which could create an uneven playing field.
For the next phase of extremely large AI models to be implemented, hardware needs to improve further to solve constraints imposed by Moore’s law.2 Smarter algorithms for training and inferencing, faster memory bandwidth for computation, and more advanced manufacturing technologies for fabricating chips are all required to improve scale and bring down costs.
AI has the potential to enhance and reshape the global economy in unpredictable ways, and is likely to be impactful to the investments we make on behalf of our clients in the coming decades. Each sector may see new contenders trying to disrupt incumbents through better solutions and cheaper cost-leveraging AI. Just as the internet has transformed print media, and smartphones the way we communicate, AI is likely to exert its influence for many years to come.
At Newton, we strive to navigate and thrive through this third epoch of tech by harnessing our multidimensional research platform that maximises the strength of our collaborative culture.
1 Source: Sizing the prize: What’s the real value of AI for your business and how can you capitalise? https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf
2 Moore’s law states that processor speeds, or overall processing power for computers, will approximately double every 18 months.
3 Source: How AI can enable a sustainable future? https://www.pwc.co.uk/sustainability-climate-change/assets/pdf/how-ai-can-enable-a-sustainable-future.pdf
This is a financial promotion. These opinions should not be construed as investment or other advice and are subject to change. This material is for information purposes only. This material is for professional investors only. Any reference to a specific security, country or sector should not be construed as a recommendation to buy or sell investments in those securities, countries or sectors. Please note that holdings and positioning are subject to change without notice. This article was written by members of the NIMNA investment team. ‘Newton’ and/or ‘Newton Investment Management’ is a corporate brand which refers to the following group of affiliated companies: Newton Investment Management Limited (NIM) and Newton Investment Management North America LLC (NIMNA). NIMNA was established in 2021 and is comprised of the equity and multi-asset teams from an affiliate, Mellon Investments Corporation.